Supertone

@notnotrishi Available for $49 USD during our opening sale.

@Supertone_ai cool! how much does a custom voice embedding download cost?

Turn your voice into a deployable, edge-native TTS. No cloud dependency. No generic voices. Just yours. Voice Builder works seamlessly with Supertonic 2, our open, edge-native TTS engine — giving you full ownership and control. - - - How it works - - - 1. Record in the browser or upload audio 2. Test your voice with our TTS system 3. Purchase once with permanent ownership 4. Download your custom voice embedding 5. Deploy on-device, anywhere Supertonic 2 runs

Voice Builder: supertonic.supertone.ai/voice-builder GitHub: github.com/supertone-inc/…

We’re seeing builders from the community move fast with Supertonic, our lightweight, on-device text-to-speech model. From browser extensions and demos to full-fledged apps, creators are already shipping with Supertonic in real-world products. 🚀 Read Aloud, PageEcho, VoiceChat … These are just a few of the projects we witnessed make waves with Supertonic. Curious how each team is using this on-device TTS in practice? Let’s take a closer look at the article! linkedin.com/feed/update/ur…

Read Aloud, trusted by more than 6 million users, has added Supertonic as a new voice option to its text-to-speech Chrome extension. 📢 The Read Aloud team praised Supertonic for its ability to deliver “amazing voices that can run on commodity devices,” and shared expectations for the voices’ adoption across the platform. Try Supertonic on Read Aloud! ⚡ Chrome extension: lnkd.in/g-s6us5G ⚡ MS Edge: lnkd.in/gRkSi8sX ⚡ TTS Tool: lnkd.in/gvGCng6W As a platform widely used in schools, Read Aloud’s adoption of Supertonic is expected to lower the barrier to high-quality text-to-speech without specialized hardware, as Supertonic is designed to deliver strong voice quality directly on-device. 📚 We’re grateful for the trust and feedback, and we’ll continue focusing on making voice AI more accessible.

@RachdiSham64061 Hi, the current model may occasionally mispronounce certain inputs (e.g., pluralized acronyms or text that hasn’t been fully normalized). We’re actively working to improve overall TTS stability, with updates planned for an upcoming release. Thank you for reporting this issue!

@Supertone_ai There is a problem with this model, as it hallucinates when reading single words and short sentences. We hope to solve this problem as soon as possible.

Now with a Python package, Supertonic fits into your workflow even better. 📂 Fresh updates: ➤ Supertonic is available as a pip-installable Python package for use in Python and CLI environments. ➤ Six new voice presets have been added, offering ten presets in total for exploring a broader range of vocal characteristics. ➤ The model checkpoint has been updated with ONNX-level optimizations that improve inference speed. Voice embedding functionality is currently in preparation and will allow developers to generate speaker embeddings for their own applications. In addition, Supertonic is provided as an open-weight model: the weights are publicly available, but the training code and data are not open source. We want to ensure this distinction is communicated clearly. 🔗 Python Package for Supertonic: supertone-inc.github.io/supertonic-py/ 🔗 Available Voices: supertone-inc.github.io/supertonic-py/… 🔗 Updates on Hugging Face Model Card: huggingface.co/Supertone/supe… 📨 For ongoing updates, the developer newsletter is also available here: subscribepage.io/W7rrGO We’re grateful for the continued interest. Stay tuned for what’s next!

@DAMSONCOIN Hi! I checked with our team on this issue. Could you please try downloading the newest trial version from our official website and authenticate with your license key? If the issue continues, please let us know. Our website URL: supertone.ai/en/clear

@notnotrishi A very neat demo you have here! We are glad you tried Supertonic out.

Supertonic by @Supertone_ai is one hell of a TTS model 👏 Low memory footprint (just 66M params), instant generation, runs literally on any device -- and the best part is it can handle financial/technical units, date times, phone numbers etc. without any extra work. (ignore the slight echo from my laptop, the model output is pretty solid)

@dtOEJU5Sp1WjHud @Shift_contents MV公開おめでとうございます！🎉 バターのったどら焼きとつやつやのあんこが美味しそうすぎて、思わずどきどきしてしまいました☺️ 🥞 すてきな作品にタグ付けしてくださり、本当にありがとうございます。

@Shift_contents @Supertone_ai ​😭【ついに...涙の完成！】😭 ​長い時間をかけて（2日30時間ぐらい?(｡･ω･｡)w）制作したオリジナルMV 『はっぴーどら焼きステップ』を公開します！ ​🌟 主人公DWの表情や、街がピンクに染まるシーンなど、こだわりを詰め込みました！ぜひ、最後まで見て感想を教えてください！ ​そして、何度も修正に付き合ってくれたGeminiにも感謝！(笑) 音声はお馴染みSUPERTONESHIFT:Lilyだよ！ youtu.be/kIxG02XCjGs?si… ​#MV投稿 #オリジナルソング #はっぴーどら焼きステップ #感動をありがとう

@lomsa005 @lomsa005 Sorry! I meant to type Lomsa*

Hi Lomas. For this model, we currently provide English TTS. We’re planning to release additional supported languages such as Korean, Japanese, Spanish, and Portuguese in API form. Also, thank you for the suggestion—it actually reminded me of a phrase my Georgian friend once taught me: “ეს არის სასიამოვნო”. (Still one of my fav thing to say randomly) We don’t have concrete plans to support it yet, but we’ll continue working to make our tools more useful in the future! Thank you.

We introduce Supertonic TTS. The world's fastest and open source model comes with more perks: > On-device by design: near-zero latency and more privacy > Blazing speed: 2,500+ cps with real-time factor down to 0.001 (RTX 4090) and 0.006 (M4 Pro) > Ultra lightweight: just 66M parameters. > No cloud dependency-runs smoothly on mobile, web, and even Raspberry Pi > Developer-friendly: you get examples in 8+ languages (C++, C#, Java, JavaScript, Rust, Go, Swift, Python) Try Supertonic and let us know your thoughts! 🔹 Live interactive demo (Hugging Face Spaces): huggingface.co/spaces/Superto…

@hibaby343 Currently, the model on Hugging face supports English TTS only, with four voices (2Female/2Male). API support for Korean, Japanese, Spanish, and Portuguese is on the way! Appreciate the interest, thank you!

@Supertone_ai How many languages and voices

@opener_ai Spine-tingling work, Thanks for tagging us for the voice part!

Ugh, that’s such a shame. I was running out of time and wasn't sure if I should even try, and I think I ended up going off-topic. Plus, I should have generated it with the watermark, but I only realized that after everything was done, so I just forced the logo into the edit line. I feel bad for Hailuo because of all the mistakes. Still, I'm focusing on the meaning of having created and submitted a story I really wanted to tell. :) I was watching Stranger Things recently and was captivated by that color palette, so I tried making this video. Even so, I wanted to deliver a short message about the horror of capitalism. Please understand that since I made this video in just two hours, the message might not be conveyed very effectively Image: Nanobananapro Video: Hailuo 100% Music: Suno Narration: Supertone @HailuoAI @Supertone_ai #HailuoHorror #aimovie

@RachdiSham64061 Thank you for sharing this suggestion — we’re still exploring various use cases, and your input is truly appreciated!

@Supertone_ai This tts model will benefit us blind people more than our dear developers imagine, but until we find a way to make it work with screen readers. Could you look into using your awesome model through Microsoft speach API 5?

@ryanator2000000 Thanks so much for trying the demo and letting us know! Could you tell us which Android device and browser you’re using so we can look into it for future improvements?

@Supertone_ai The live demo produces unintelligible noise when I try it on my phone (android). It works great on my laptop though. Any idea why it doesn't work on my phone?

@RodFdw @xenovacom So far, yes! However, we are thinking of offering more language options in the future, maybe in API format

@xenovacom @Supertone_ai English only?

Introducing Supertonic WebGPU: blazingly fast text-to-speech running 100% locally in your browser. ⚡️ Generate free audiobooks in seconds, no installation required. In this example, I created a 5-hour audiobook of The Great Gatsby in less than 3 minutes! 🤯 Try it yourself! 👇

Massive kudos to the @Supertone_ai team for this fantastic model release. I honestly can't believe something like this is finally possible! 🤩 🔗 Demo + source code: huggingface.co/spaces/webml-c…